image_classification_for_fracture

This model is a fine-tuned version of google/vit-base-patch16-224-in21k on the imagefolder dataset. It achieves the following results on the evaluation set:

• Loss: 0.4783
• Accuracy: 0.85

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 5e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 64
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• lr_scheduler_warmup_ratio: 0.1
• num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Accuracy
No log	0.8	2	0.6696	0.75
No log	2.0	5	0.6296	0.7
No log	2.8	7	0.5853	0.775
0.639	4.0	10	0.5731	0.8
0.639	4.8	12	0.5430	0.825
0.639	6.0	15	0.5223	0.85
0.639	6.8	17	0.5036	0.8
0.5453	8.0	20	0.4783	0.85

Framework versions

• Transformers 4.39.1
• Pytorch 2.2.1+cpu
• Datasets 2.18.0
• Tokenizers 0.15.2